diff options
| author | Loren Merritt <lorenm@u.washington.edu> | 2006-10-01 21:25:17 +0000 |
|---|---|---|
| committer | Loren Merritt <lorenm@u.washington.edu> | 2006-10-01 21:25:17 +0000 |
| commit | 2833fc46462abd841e3d69242be9919c98ad3696 (patch) | |
| tree | 4e9513d1913ff238c35e066877c8795b40d561c8 /libavcodec/i386/dsputil_mmx_avg.h | |
| parent | e0769997cb753d8a3c9039b1de1542987cfc0692 (diff) | |
approximate qpel functions: sacrifice some quality for some decoding speed. enabled on B-frames with -lavdopts fast.
Originally committed as revision 6412 to svn://svn.ffmpeg.org/ffmpeg/trunk
Diffstat (limited to 'libavcodec/i386/dsputil_mmx_avg.h')
| -rw-r--r-- | libavcodec/i386/dsputil_mmx_avg.h | 48 |
1 files changed, 48 insertions, 0 deletions
diff --git a/libavcodec/i386/dsputil_mmx_avg.h b/libavcodec/i386/dsputil_mmx_avg.h index 8692fb0487..6dedc57647 100644 --- a/libavcodec/i386/dsputil_mmx_avg.h +++ b/libavcodec/i386/dsputil_mmx_avg.h @@ -818,3 +818,51 @@ static void DEF(avg_pixels16_xy2)(uint8_t *block, const uint8_t *pixels, int lin DEF(avg_pixels8_xy2)(block+8, pixels+8, line_size, h); } +#define QPEL_2TAP_L3(OPNAME) \ +static void DEF(OPNAME ## 2tap_qpel16_l3)(uint8_t *dst, uint8_t *src, int stride, int h, int off1, int off2){\ + asm volatile(\ + "1: \n\t"\ + "movq (%1,%2), %%mm0 \n\t"\ + "movq 8(%1,%2), %%mm1 \n\t"\ + PAVGB" (%1,%3), %%mm0 \n\t"\ + PAVGB" 8(%1,%3), %%mm1 \n\t"\ + PAVGB" (%1), %%mm0 \n\t"\ + PAVGB" 8(%1), %%mm1 \n\t"\ + STORE_OP( (%1,%4),%%mm0)\ + STORE_OP(8(%1,%4),%%mm1)\ + "movq %%mm0, (%1,%4) \n\t"\ + "movq %%mm1, 8(%1,%4) \n\t"\ + "add %5, %1 \n\t"\ + "decl %0 \n\t"\ + "jnz 1b \n\t"\ + :"+g"(h), "+r"(src)\ + :"r"((long)off1), "r"((long)off2),\ + "r"((long)(dst-src)), "r"((long)stride)\ + :"memory"\ + );\ +}\ +static void DEF(OPNAME ## 2tap_qpel8_l3)(uint8_t *dst, uint8_t *src, int stride, int h, int off1, int off2){\ + asm volatile(\ + "1: \n\t"\ + "movq (%1,%2), %%mm0 \n\t"\ + PAVGB" (%1,%3), %%mm0 \n\t"\ + PAVGB" (%1), %%mm0 \n\t"\ + STORE_OP((%1,%4),%%mm0)\ + "movq %%mm0, (%1,%4) \n\t"\ + "add %5, %1 \n\t"\ + "decl %0 \n\t"\ + "jnz 1b \n\t"\ + :"+g"(h), "+r"(src)\ + :"r"((long)off1), "r"((long)off2),\ + "r"((long)(dst-src)), "r"((long)stride)\ + :"memory"\ + );\ +} + +#define STORE_OP(a,b) PAVGB" "#a","#b" \n\t" +QPEL_2TAP_L3(avg_) +#undef STORE_OP +#define STORE_OP(a,b) +QPEL_2TAP_L3(put_) +#undef STORE_OP +#undef QPEL_2TAP_L3 |